in Fig. 10.5 notyetdiscussed: the integration of concepts withone's f-knowledge base. Suppose I say toyou:I talked to
David Wasserstein today. In isolation, the proper nameDavid Wassersteinpurports to refer; you assume that I intended it
actuallytorefer. Ifyouknow David Wasserstein(i.e.heis part of yourknowledge base),you canestablish theintended
reference in your own construal of the world. If you do not know him, the proper name is unsatisfiedfor you.
More generally, four different situations can arise.
(18)a. The purported referent is presentinyour f-knowledge base or thereadily availablecontext, and it satisfies the
referential expression.
b. The purported referent is not present in your f-knowledge base or the readily available context, and so you
add a new character into your f-knowledge base to satisfy the referential expression. If I anticipate this
situation for you, I will use a specific indefinite expression likea friend of mineorthis friend of mine, David
Wassersteinrather than a definite description or an unadorned proper name.
c. The purported referent is in conflict with your f-knowledge base, as in Russell's famous examplethe present
king of France. Or it is in conflict with the readily availablecontext, for example in a situation when I speak of
the apple on the tableand you see either no apples or two apples on the table. In this case you have to fall back
on some repertoire of repair strategies (H. H. Clark 1996: ch. 9): guessing what I intend,^168 asking me for
clarification, deciding to ignore me altogether, and so on.
d. The purported referent contains descriptive features that inherently conflict with each other, so that nothing
can possibly satisfy the expression. In such a case, for instancethe square circle, youfind the expression
anomalous, and must again fall back on repair strategies.This is all clear with the reference of NPs. We next turn to the reference of sentences. The standard custo min the
formal semantics tradition, going back to Frege (1892), is that the intended referent of a (declarative) sentence is a
truth value. I must confess I have never understood the argument for this position (e.g. Chierchia and McConnell-
Ginet 1990: ch. 2); to unravel it is beyond the scope of the present text (see however the discussion in Seuren 1998:
375ff.). I
REFERENCE AND TRUTH 325
(^168) This would include the well-known situation described by Donnellan (1966) , here adapted to the conceptualist approach: Joan speaks to Ida ofthe man drinking a martini
and gestures toward Edgar. Ida, however, happens to know/believethat Edgar is drinking water,and hence cannot take hi mto satisfy Joan's description. Nevertheless, Ida
will probably be charitable and understand Joan to intend the phrase to refer to Edgar. We do not have to worry about whether the phrasereally refers, only about how
language users treat it.